Eddies are of great concern to ship operations and have an important influence on heat transfer in the sea. In this article, a highly concurrent and accurate method incorporating the winding angles and density-based clustering (WA-DBC) algorithms is proposed for automatic eddy detection from surface current maps generated by high-frequency radars (HFRs). As a case study, eddies and their characteristics, such as centers, radii, and lifespans, were extracted using both the WA-DBC and vector geometry algorithms from the HFR surface current maps of the north of Qiongzhou Strait, South China Sea. The eddy trajectories obtained show that most eddies in this region during the experiment period had a radius of 3–30 km and negative correlation existed between the radius and the life span (i.e., eddies with a smaller radius survived a longer lifespan). Furthermore, most of the eddies appeared in the central part of the study area and lasted for 40–100 min. This study shows that HFR is useful for detecting submesocale eddies for research works and applications in marine observation.

中文翻译：

---

使用高频雷达进行亚中尺度涡流观测：以南海北部为例

涡流是船舶运营中非常关注的问题，并且对海洋中的热传递有重要影响。在本文中，提出了一种高度并发且准确的方法，该方法结合了绕组角和基于密度的聚类（WA-DBC）算法，用于从高频雷达（HFR）生成的表面电流图中自动进行涡流检测。作为案例研究，利用WA-DBC和矢量几何算法从南海琼州海峡北部的HFR表面电流图中提取了涡旋及其特征，例如中心，半径和寿命。获得的涡旋轨迹表明，该区域在实验期间的大多数涡旋半径为3–30 km，并且半径与寿命之间存在负相关关系（即，半径较小的涡旋寿命更长）。此外，大多数涡流出现在研究区域的中部，持续了40-100分钟。这项研究表明，HFR可用于检测用于海洋研究的研究工作和应用中的近中生涡流。